Princeton ad blocker uses optical character recognition and word detection to find ads

Jason Koebler, writing for Motherboard:

A team of Princeton and Stanford University researchers (Arvind Narayanan, Dillon Reisman, Jonathan Mayer, and Grant Storey) has fundamentally reinvented how ad-blocking works. [...] The ad blocker [...] evaded anti ad-blocking scripts on 50 out of the 50 websites it was tested on, and can block Facebook ads that were previously unblockable.

Jason adds that the techniques employed include "optical character recognition, design techniques, and container searches (the boxes that ads are commonly put in on a page) to detect words like "sponsored" or "close ad" that are required to appear on every ad." 

Even without these specialized techniques, ad blocker usage grew 30% last year (source: PageFair). That's how motivated people are to clean up the garbage pit of an experience many websites are because of intrusive and obnoxious ads.

Assuming continued favorable legislation that protects consumer interests, for example, requiring ads to be clearly marked and to be understood as ads, then we can assume two things:

  • Ad blocking usage will continue to grow as desperate publishers perversely place more ads on their sites as revenues fall.
  • Ad blocking technology will continue to get better, and will match, if not exceed the ability of a human to recognize ads.